# Jailbreaking Text-to-Image Models Through Cracks: Navigating Heterogeneous Safety Filters via Multi-Agent Debate

arXiv: https://arxiv.org/abs/2609.01168 · HTML: https://mlsanity.com/paper/2609.01168/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Agent threats, guardrails and governance, one of 16 directions in LLM agents and tool use.
- Builds on 36 of its 48 references that are in the map.
- First posted on arXiv 1 Sep 2026.

## Placed in

- LLM agents and tool use / [Agent threats, guardrails and governance](https://mlsanity.com/agents/agent-threats-guardrails-and-governance/)
- LLM safety and security / [Vision-language and text-to-image safety](https://mlsanity.com/safety/vision-language-and-text-to-image-safety/)

## Builds on (most cited first, 30 of 36)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis](https://mlsanity.com/paper/2307.01952/) (2023)
- [CLIPScore: A Reference-free Evaluation Metric for Image Captioning](https://mlsanity.com/paper/2104.08718/) (2021)
- [Improving Factuality and Reasoning in Language Models through Multiagent Debate](https://mlsanity.com/paper/2305.14325/) (2023)
- [Jailbreaking Black Box Large Language Models in Twenty Queries](https://mlsanity.com/paper/2310.08419/) (2023)
- [Red Teaming Language Models with Language Models](https://mlsanity.com/paper/2202.03286/) (2022)
- [Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debate](https://mlsanity.com/paper/2305.19118/) (2023)
- [Tree of Attacks: Jailbreaking Black-Box LLMs Automatically](https://mlsanity.com/paper/2312.02119/) (2023)
- [Erasing Concepts from Diffusion Models](https://mlsanity.com/paper/2303.07345/) (2023)
- [Safe Latent Diffusion: Mitigating Inappropriate Degeneration in Diffusion Models](https://mlsanity.com/paper/2211.05105/) (2022)
- [On the Number of Linear Regions of Deep Neural Networks](https://mlsanity.com/paper/1402.1869/) (2014)
- [Red-Teaming the Stable Diffusion Safety Filter](https://mlsanity.com/paper/2210.04610/) (2022)
- [Ring-A-Bell! How Reliable are Concept Removal Methods for Diffusion Models?](https://mlsanity.com/paper/2310.10012/) (2023)
- [Rainbow Teaming: Open-Ended Generation of Diverse Adversarial Prompts](https://mlsanity.com/paper/2402.16822/) (2024)
- [MMA-Diffusion: MultiModal Attack on Diffusion Models](https://mlsanity.com/paper/2311.17516/) (2023)
- [SneakyPrompt: Jailbreaking Text-to-image Generative Models](https://mlsanity.com/paper/2305.12082/) (2023)
- [Unsafe Diffusion: On the Generation of Unsafe Images and Hateful Memes From Text-To-Image Models](https://mlsanity.com/paper/2305.13873/) (2023)
- [Prompting4Debugging: Red-Teaming Text-to-Image Diffusion Models by Finding Problematic Prompts](https://mlsanity.com/paper/2309.06135/) (2023)
- [Can Machines Help Us Answering Question 16 in Datasheets, and In Turn Reflecting on Inappropriate Content?](https://mlsanity.com/paper/2202.06675/) (2022)
- [Should we be going MAD? A Look at Multi-Agent Debate Strategies for LLMs](https://mlsanity.com/paper/2311.17371/) (2023)
- [Improving Image Captioning with Better Use of Captions](https://mlsanity.com/paper/2006.11807/) (2020)
- [Controllable Generation with Text-to-Image Diffusion Models: A Survey](https://mlsanity.com/paper/2403.04279/) (2024)
- [Actor-Critic Reinforcement Learning for Control with Stability Guarantee](https://mlsanity.com/paper/2004.14288/) (2020)
- [AutoRedTeamer: Autonomous Red Teaming with Lifelong Attack Integration](https://mlsanity.com/paper/2503.15754/) (2025)
- [Perception-guided Jailbreak against Text-to-Image Models](https://mlsanity.com/paper/2408.10848/) (2024)
- [DreamStyler: Paint by Style Inversion with Text-to-Image Diffusion Models](https://mlsanity.com/paper/2309.06933/) (2023)

## Abstract (arXiv)

Text-to-image (T2I) models remain vulnerable to jailbreak attacks that elicit Not-Safe-For-Work (NSFW) content, despite increasingly being guarded by heterogeneous, multi-layer safety stacks combining text filters, image classifiers, and cross-modal detectors. Existing jailbreak studies either optimize against individual filters or query the complete pipeline with aggregate feedback, making it difficult to identify the active constraint and adapt to conflicts across safety layers. In this paper, we introduce the Detection Surface, a unified geometric framework that characterizes the decision boundaries induced by heterogeneous T2I safety filters and their joint effect on the jailbreak search space. This formulation reveals that successful evasion is governed by a sparse and non-convex region shaped by cross-layer conflicts, where mutations that bypass one filter may increase exposure to another. Motivated by this analysis, we propose CRACK, a multi-agent debate framework for adaptive jailbreak search that decomposes jailbreak search into exploration, diagnosis, and arbitration. CRACK coordinates an Attack Agent, a Defense Agent, and a Judge Agent to iteratively generate prompt mutations, obtain layer-specific diagnostic feedback, and optimize mutation strategies through reward-guided refinement. Through repeated rounds of debate, CRACK adapts its search direction to the evolving cross-layer constraints while preserving the original harmful intent. Extensive experiments across multiple T2I models, datasets, and safety configurations show that CRACK achieves Attack Success Rates (ASR) of up to 99.63% under composite defenses, while requiring fewer queries than existing methods and maintaining semantic fidelity.